cryptopals/crime_notes.txt at master · AdrS/cryptopals · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
#Dealing With Huffman Coding
##Two Tries
padding = genRandomPadding()
'https://victim.com/?secret=' + known + guess + padding
'https://victim.com/?secret=' + known + padding + guess

Should have same size after huffman encoding because both have same characters
different size => must be because of LZ77

ex:
'https://victim.com/?secret=52427{}{}{}{}{}' and
'https://victim.com/?secret=5242{}{}{}{}{}7'

##Character Set Pool

for guess in alphabet
	alphabet_with_guess = ''.join(sorted(list(set(alphabet) - {guess})))

'https://victim.com/?secret=' + known + guess + padding + alphabet_with_guess

Character frequencies the same for each request => differences in size should be from LZ77

alphabet = string.digits
Guess 7: 'https://victim.com/?secret=52427{}{}{}{}{}012345689'
Guess 8: 'https://victim.com/?secret=52428{}{}{}{}{}012345679'


#Conflict Recovery

-make guesses 2 characters at a time
Pros: reliable, cons: expensive

Secret could compress well against itself ==> :(
	Look at compression ratios of guesses to identify this